Search

Abstract Background Amyotrophic Lateral Sclerosis (ALS) is a highly heterogenous neurodegenerative disorder that primarily affects upper and lower motor neurons, affecting additional cell types and brain regions. Underlying molecular mechanisms are still elusive, in part due to disease heterogeneity. Molecular disease subtyping through integrative analyses including RNA editing profiling is a novel approach for identification of molecular networks involved in pathogenesis. Methods We aimed to highlight the role of RNA editing in ALS, focusing on the frontal cortex and the prevalent molecular disease subtype (ALS-Ox), previously determined by transcriptomic profile stratification. We established global RNA editing (editome) and gene expression (transcriptome) profiles in control and ALS-Ox cases, utilizing publicly available RNA-seq data (GSE153960) and an in-house analysis pipeline. Functional annotation and pathway analyses identified molecular processes affected by RNA editing alterations. Pearson correlation analyses assessed RNA editing effects on expression. Similar analyses on additional ALS-Ox and control samples (GSE124439) were performed for verification. Targeted re-sequencing and qRT-PCR analysis targeting CACNA1C, were performed using frontal cortex tissue from ALS and control samples (n = 3 samples/group). Results We identified reduced global RNA editing in the frontal cortex of ALS-Ox cases. Differentially edited transcripts are enriched in synapses, particularly in the glutamatergic synapse pathway. Bioinformatic analyses on additional ALS-Ox and control RNA-seq data verified these findings. We identified increased recoding at the Q621R site in the GRIK2 transcript and determined positive correlations between RNA editing and gene expression alterations in ionotropic receptor subunits GRIA2, GRIA3 and the CACNA1C transcript, which encodes the pore forming subunit of a post-synaptic L-type calcium channel. Experimental data verified RNA editing alterations and editing-expression correlation in CACNA1C, highlighting CACNA1C as a target for further study. Conclusions We provide evidence on the involvement of RNA editing in the frontal cortex of an ALS molecular subtype, highlighting a modulatory role mediated though recoding and gene expression regulation on glutamatergic synapse related transcripts. We report RNA editing effects in disease-related transcripts and validated editing alterations in CACNA1C. Our study provides targets for further functional studies that could shed light in underlying disease mechanisms enabling novel therapeutic approaches.

Abstract Background Amyotrophic lateral sclerosis (ALS) is a rare progressive neurodegenerative disease that affects upper and lower motor neurons. As the molecular basis of the disease is still elusive, the development of high-throughput sequencing technologies, combined with data mining techniques and machine learning methods, could provide remarkable results in identifying pathogenetic mechanisms. High dimensionality is a major problem when applying machine learning techniques in biomedical data analysis, since a huge number of features is available for a limited number of samples. The aim of this study was to develop a methodology for training interpretable machine learning models in the classification of ALS and ALS-subtypes samples, using gene expression datasets. Methods We performed dimensionality reduction in gene expression data using a semi-automated preprocessing systematic gene selection procedure using Statistically Equivalent Signature (SES), a causality-based feature selection algorithm, followed by Boosted Regression Trees (XGBoost) and Random Forest to train the machine learning classifiers. The SHapley Additive exPlanations (SHAP values) were used for interpretation of the machine learning classifiers. The methodology was developed and tested using two distinct publicly available ALS RNA-seq datasets. We evaluated the performance of SES as a dimensionality reduction method against: (a) Least Absolute Shrinkage and Selection Operator (LASSO), and (b) Local Outlier Factor (LOF). Results The proposed methodology achieved 85.18% accuracy for the classification of cerebellum or frontal cortex samples as C9orf72-related familial ALS, sporadic ALS or healthy samples. Importantly, the genes identified as the most determinative have also been reported as disease-associated in ALS literature. When tested in the evaluation dataset, the methodology achieved 88.89% accuracy for the classification of sporadic ALS motor neuron samples. When LASSO was used as feature selection method instead of SES, the accuracy of the machine learning classifiers ranged from 74.07 to 96.30%, depending on tissue assessed, while LOF underperformed significantly (77.78% accuracy for the classification of pooled cerebellum and frontal cortex samples). Conclusions Using SES, we addressed the challenge of high dimensionality in gene expression data analysis, and we trained accurate machine learning ALS classifiers, specific for the gene expression patterns of different disease subtypes and tissue samples, while identifying disease-associated genes.

Abstract Background YKL-40 (also known as Chitinase 3-like 1) is a glycoprotein produced by inflammatory, cancer and stem cells. Its physiological role is not completely understood but YKL-40 is elevated in the brain and cerebrospinal fluid (CSF) in several neurological and neurodegenerative diseases associated with inflammatory processes. Yet the precise characterization of YKL-40 in dementia cases is missing. Methods In the present study, we comparatively analysed YKL-40 levels in the brain and CSF samples from neurodegenerative dementias of different aetiologies characterized by the presence of cortical pathology and disease-specific neuroinflammatory signatures. Results YKL-40 was normally expressed in fibrillar astrocytes in the white matter. Additionally YKL-40 was highly and widely expressed in reactive protoplasmic cortical and perivascular astrocytes, and fibrillar astrocytes in sporadic Creutzfeldt-Jakob disease (sCJD). Elevated YKL-40 levels were also detected in Alzheimer’s disease (AD) but not in dementia with Lewy bodies (DLB). In AD, YKL-40-positive astrocytes were commonly found in clusters, often around β-amyloid plaques, and surrounding vessels with β-amyloid angiopathy; they were also distributed randomly in the cerebral cortex and white matter. YKL-40 overexpression appeared as a pre-clinical event as demonstrated in experimental models of prion diseases and AD pathology. CSF YKL-40 levels were measured in a cohort of 288 individuals, including neurological controls (NC) and patients diagnosed with different types of dementia. Compared to NC, increased YKL-40 levels were detected in sCJD (p 0.05, AUC = 0.71) or in DLB/Parkinson’s disease dementia (PDD) (p > 0.05, AUC = 0.70). Further, two independent patient cohorts were used to validate the increased CSF YKL-40 levels in sCJD. Additionally, increased YKL-40 levels were found in genetic prion diseases associated with the PRNP-D178N (Fatal Familial Insomnia) and PRNP-E200K mutations. Conclusions Our results unequivocally demonstrate that in neurodegenerative dementias, YKL-40 is a disease-specific marker of neuroinflammation showing its highest levels in prion diseases. Therefore, YKL-40 quantification might have a potential for application in the evaluation of therapeutic intervention in dementias with a neuroinflammatory component.